The Vestibular Navigation Hypothesis: A Progress Report

The hypothesis that the vestibular system can function as an inertial guidance system for animals’ navigation across terrain is an old but still incompletely established idea (eg., Exner, 1893; for a recent review see Potegal, 1982). In this “dead-reckoning” or “path integration” hypothesis, velocity signals from the semicircular canals and the otolith organs are integrated in the central nervous system to provide the organism with information about its respective angular and linear displacements from some origin. The plausibility of this hypothesis depends upon the demonstration that the nervous system has the capability for such computations. There is, in fact, neurophysiological evidence for at least two integrators operating on vestibular input within the oculomotor control systems. One of these, possibly located within the penabducens nucleus, provides a steady-state, gaze-related input to the eye muscles (Robinson, 1974). A second, “velocity storage” integrator is thought to generate the signal for postrotatory and optokinetic afternystagmus (Raphan, et al, 1979). Recent evidence suggests that input from the otolith organs exerts a strong control over the time constant of this latter integrator (Waespe, et al, 1985), a proposal which is consistent with some behavioral observations made by Cohen and Potegal (see below). While the existence of an integrator specifically subserving vestibular navigation has yet to be demonstrated, circumstantial evidence suggests that the basal ganglia may be involved in such computations (Abraham, et al, 1983). In any event, the nervous system clearly has this general computational capability.